DU
PR O
RE
TO
GA L IL LE IS IT
20
TI
rior/mediocre performance. All else being equal, the most efficient trader will enjoy the highest net returns. In addition, monitoring a manager’s traded portfolio (the returns associated with the stocks purchased or sold during the quarter) may also provide foresight as to the overall portfolio’s future performance. Further, to the extent plan sponsors establish policies that directly affect the trading of their assets (i.e., directed brokerage, transitions or asset rebalancing, and manager selections) it is doubly important that they do so in a fiscally responsible manner.
AR
P
CE
STEVEN GLASS is managing director of the board of the Plexus Group, Inc. in Los Angeles, CA.
lexus has long been on record that superior performance requires more than just good stock picking; the manager must also be able to trade those picks efficiently. We see too many instances where unnecessary trading costs offset superior stock picks, resulting in sub-par net returns.1 Unfortunately, these two tasks typically constitute separate and distinct processes within each management firm. Underperformance by either can result in weak overall performance. As a consequence, both regulatory agencies and fiscal sagacity compel the periodic review of trading costs. For plan sponsors this is more than just a technical matter of satisfying procedural due diligence. As Andre Perold has frequently remarked, transaction costs are easier to predict and control than returns on stocks. At the heart of this observation is a recognition that managers typically employ consistent trading strategies over time. Consequently, as long as investment mandates remain the same (e.g., small cap growth or large cap value) their level and pattern of trading costs are largely preordained. For example, a small cap growth manager who employs a trading strategy of “averaging in” (breaking up large orders to be traded over multiple days in order to minimize Market Impact costs) will typically experience large timing/delay costs. Understanding the cost profile of a manager’s trading process provides the plan sponsor insight into the likelihood of continued supe-
IS
is chairman of the board of the Plexus Group, Inc. in Los Angeles, CA.
TH
WAYNE H. WAGNER
CL E
IN
AN Y
WAYNE H. WAGNER AND STEVEN GLASS
FO RM AT
What Every Plan Sponsor Needs to Know About Transaction Costs
THE IMPORTANCE OF TRANSACTION COST ANALYSIS When Commissions Average Only Five or Six Cents Per Share
If the total cost of trading were only 5 or 6 cents per share, there would indeed be little reason for worry. However, like the tip of an iceberg, commissions represent but a small fraction of the true costs at risk. Understanding trading costs begins with a recognition that not all trades are alike. They differ in ease or difficulty of execution. It follows that the more difficult a trade, the more expensive it will be to execute. What is less obvious (but no less important) is that these costs represent an aggregation of several distinct components. The most comprehensive view of execution costs identifies four components, albeit
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
often under different names. For purposes of clarity and common understanding, we label these components: Commissions, Market Impact, Delay Costs, and Opportunity. Brief descriptions follow below. Commissions. Commissions are the explicit charge by a broker to handle a trade. While it is easy to observe and measure, there are some nuances worth noting. In particular, different broker-services carry different commission price tags: a broker may charge only 3 cents per share for easy “vanilla” trades but 15 cents for higher risk principal type trades. Market impact. Market Impact cost represents the change in a stock’s price caused by a broker attempting to execute a trade received from a manager. It is measured by comparing the price of a stock at the point in time when the broker first received an order (from the manager) to the actual price ultimately paid for that security. This cost component has a variety of labels, such as “broker impact,” “brokerage,” “daily impact,” or just “impact.” Delay. Delay costs represent the change in a stock’s price that occurs once the manager makes a decision to buy or sell a stock, but before releasing it to a specific broker. This is sometimes referred to as “working” an order (e.g., while the manager waits for the “best” time to trade). The larger the order, the greater the likelihood of delays as the trader waits for liquidity to develop. It is measured by comparing the price of the stock at the time the manager first decides to buy or sell until the time the manager actually places the order with a broker. Opportunity. Opportunity costs represent the cost of not trading or only partially completing a desired order. It is a measure of the stock appreciation forgone by the manager as a result of not trading the entire order, and is calculated by multiplying the price appreciation of the stocks, by the percentage of stocks not traded. For several years, Plexus has published “The Iceberg of Transaction Costs.” The Iceberg graphically sets out the cost levels of each component, based on all of the trading activity tracked in the Plexus databases.2 As shown in Exhibit 1, as of June 2000, the average commission was -12 bp (4.7 cents). The average Market Impact cost was -27 bp, and the average Manager Delay cost was -62 bp. Opportunity cost averaged -16 bp on just 10% of unexecuted trades. Even excluding costs associated with not trading (i.e., Opportunity costs), managers can therefore expect to pay, on average, over 1% (and, depending on their investment mandate, as much as 4%). Arguments that these costs have minimal effect on overall portfolio returns SPRING 2001
EXHIBIT 1 The Iceberg of Transaction Costs Commission 4.7 ¢ (12 bp)
Impact 10.7¢ (27 bp)
Delay 24.7¢ (62 bp) Missed Trades 8 ¢ (16 bp)
(since they are only incurred on the traded stocks, not the entire portfolio) can be easily dismissed. Given that most managers average annual turnover rates are between 50%100% per annum, trading costs will drag overall portfolio performance down by a similar percentage. And for those managers who employ a high turnover strategy (sometimes averaging well over 150% annual turnover) the potential effect on returns will be magnified. Put these cost levels into perspective. According to the Piper data base universe of equity managers, the long-term difference in investment performance between the Top Quartile and Median equity manager was less than 3%.3 This spread holds true even across subcategories. For instance, the difference between the Top Quartile and Median growth manager was only 2.6%, while the difference for value managers was 1.6%. When comparing comingled managers, the difference fell to 1.1%. In other words, if a growth manager picked stocks that appreciated 25% but paid 2.5% in execution costs, the manager’s performance would fall from first quartile to fourth.4 Clearly, given the small differences between superior and mediocre performance, no manager or plan sponsor can afford inefficient executions. What Is “Best Execution”?
Historically, “Best Execution” was typically defined as “finding the best price at the time the trade must be TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
21
made.” While seemingly above reproach, upon closer inspection this concept is neither particularly insightful nor correct from a bottom-line plan sponsor perspective. In essence, such a philosophy refers to trading practices measured in minutes rather than days. Unfortunately, given the reality that managers properly have both the discretion and the need to “work” their orders over multiple days, plan sponsors and managers are better served viewing the implementation process in a broader context. Quite frankly, limiting the assessment of execution quality to a comparison of the time immediately surrounding the trade, encourages practices such as “averaging in” and “scaling down” (a similar practice for large sell orders). Managers are also incented to follow certain maxims such as: don’t set a new high, don’t trade at the high of the day, or don’t trade at a price higher than the Volume Weighted Average Price. Although these strategies may be appropriate in some situations, they are not in others. The focus of such an approach is to ensure that the manager doesn’t pay the worst price of that day, but ignores the very real possibility that today’s worst price may be better than tomorrow’s best price! While this concern may not be relevant to measuring the skill of a broker (who typically trades on a minuteby-minute basis), it is absolutely relevant to managers who work their orders over several days, and to the plan sponsors whose assets they invest. When considering the four cost components, it should be noted that it is fairly easy for an astute manager to minimize any single component if unconcerned about the others. For instance, Market Impact can be minimized by buying only stocks whose prices are falling, or selling stocks that rise. Unfortunately, since managers typically buy stocks on the basis of information or recommendations, limiting trades to only those stocks that fall would invariably result in buying only the stocks for which the information or recommendation was misleading. As a consequence, Opportunity costs would likely exceed the savings in Market Impact. Note that simplistic trading measures which look only at commissions and marketimpact could develop strongly misleading conclusions here. It is the manager’s job as a fiduciary to orchestrate its trade strategies to minimize the aggregate costs from all four components. Balancing these costs ultimately defines excellence in the implementation of investment ideas. “Best Execution” then, becomes more than just reducing commissions or Market Impact. Rather, “Best Execution” is that execution which delivers the maximum value of the buy/sell decision into the portfolio.5 22
Conceptually, this view is consistent with the position of the Securities and Exchange Commission’s Release No. 23170 that stated that to satisfy its obligation to obtain “best execution” a “money manager must: execute securities transactions for clients in such a manner that the client’s total cost or proceeds in each transaction is the most favorable under the circumstances” (emphasis added). Further, the SEC clarified that in making this assessment “the determinative factor is not the lowest possible commission cost but whether the transaction represents the best qualitative execution for the managed account.” What Are My Legal Obligations?
As prudent fiduciaries, plan sponsors have an obligation to monitor their managers to ensure receipt of “best execution.” In particular, ERISA Tech Rel 86-1 states that Sections 403 and 404 of the Employee Retirement Income Security Act of 1974 (ERISA) require plan sponsors to exercise “oversight authority to periodically review the investment manager’s performance… and ensure that the manager has secured the best execution of the plan’s brokerage transactions.”6 In addition to monitoring for “best execution” plan sponsors should ensure that managers comply with investment guidelines and policies. Specifically, Part II Item 12B of Form ADV requires that the manager disclose all factors used to select brokers, the reasonableness of their commissions, information regarding soft dollar arrangements, and any procedures used in directing brokerage to particular brokers in return for products and services. The accompanying Commentary to form ADV states that these disclosures “are intended to assist clients in evaluating any conflicts of interest inherent in the adviser’s arrangements for allocating brokerage.” Many managers use generic boiler-plate language with little practical value when completing Form ADV, others take this obligation seriously and provide substantive insightful information. Plan sponsors are well served to review these materials as part of their standard monitoring procedures. Further, where material information is missing, nothing prohibits a plan sponsor from requesting that data directly from the manager.7 A recent Advisory Council of the Department of Labor noted that the fiduciary obligations imposed under ERISA require plan fiduciaries to closely monitor plan expenses; including the commissions involved in soft dollar and commission recapture programs. As a consequence, the report recommends that plan sponsors be required to:
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
• List all fees greater than $5,000 paid with directed brokerage; • Certify that they are in compliance with ERISA Technical Release 86-1; • Hire only consultants with no financial arrangements with brokerage firms (to avoid conflicts of interest); • Hire independent accounting firms to audit their managers’ soft dollar disclosures; • Obtain copies of their managers and brokers projected soft dollar and directed brokerage budgets; • Manager and consultant contracts should require written disclosure of potential conflicts of interest; • Consultants should be required to disclose all compensation they receive from managers; • Managers should be required to disclose how they pay for research and how the plan sponsor benefited from that research; • Consultants and managers should be required to acknowledge their fiduciary status in writing; and • Plan sponsors should establish specific guidelines for soft dollar and directed brokerage programs. Most recently, the SEC’s Office of Compliance, Inspections and Examinations released their Inspection Report on the Soft Dollar Practices of Broker-Dealers, Investment Advisers and Mutual Funds. The report noted that broker-dealers and investment advisers were “not consistently applying the standards articulated in the 1986 Release” and therefore issued a number of recommendations to the commission including: • Revise Form ADV to improve disclosure of the products and services purchased with soft dollars; • Managers should be required to provide more detailed itemization of soft dollar purchases, including total commission commitments and total expenses on a client-specific basis; and • Revise Form ADV to require disclosure of availability of commission recapture if any of the manager’s clients directs brokerage. While regulations representing the consensus of all governing bodies are still in a state of flux, the general outlines are similar: costs are important and plan sponsors are obliged to pay attention.
SPRING 2001
POTENTIAL ISSUES AND PROBLEMS
Transaction cost analysis has two inter-related goals: identification of abuses, and evaluation of inefficient trading processes. The first goal, while certainly relevant to overall trading efficiency, also represents a stand-alone obligation for fiduciaries. To the extent abuses are discovered, plan sponsors who fail to act become themselves complicit in the continued wrongdoing. The second goal, systematic inefficiencies, while perhaps not rising to the level of a breach of fiduciary responsibility, often costs many funds more than isolated incidents of abuse. In both cases, plan sponsors are well served to review their managers’ trading practices. On a practical level this boils down to several specific areas of inquiry. Are My Managers' Executions Consistent with “Best Execution” and Up to Par Compared to Similar Trades?
If nothing else, plan sponsors should be able to answer two questions regarding their managers’ execution efficiency: 1. How much was paid out of the assets of the fund to buy and sell the stocks in the portfolio? 2. Were those costs reasonable? This sounds straightforward, and it is. However, a couple of nuances are worth noting. First, plan sponsors must be certain that they are in fact calculating the total execution costs being incurred. Referring to the earlier discussion on cost components, many measurement systems limit their analysis to commissions and Market Impact while ignoring Delay costs. Unfortunately, as also described earlier, such practices can lead to misleading conclusions as to the actual size of the transaction costs, as well as the appropriateness of a particular trade strategy. Bottom line: without knowing the total costs paid to implement a trade, it is impossible to put any cost component into context. Plan sponsors therefore have no way of knowing whether their managers were being “penny wise and pound foolish.” Aside from calculating the entire transaction costs, plan sponsors must also be certain they are using a relevant benchmark. In order to get a true benchmark of expected costs (against which a manager’s actual costs can be contrasted), it is important to take into account trade difficulty: the characteristics of what was being traded and the market environment into which the trade was released. TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
23
Size of the company. Small companies tend to have fewer institutional investors holding their stock, which in turn makes it less likely for a manager to find a comparable order on the other side of its desired trade. As a consequence, managers must either pay a broker significant premiums to take the “other side” itself, or put together enough contra orders to take the other side. Urgency. Orders that require instant liquidity (i.e., must be completed regardless of availability of another institutional investor willing to take the other side) often employ Market Makers who will provide the needed liquidity, but only at a price which affords them an expected profit. So for example, a growth or momentum manager who bases his stock selections on news breaks or price momentum will almost always have to outbid other managers trying to trade the same securities. In contrast, a value manager (who selected stocks perceived to have undiscovered value) will often encounter situations in which he alone is seeking to buy or sell a particular security. Whoever is providing “liquidity” to the market can experience favorable trading situations in which it is possible to trade at a profit. Indeed, failure to profit from these opportunities might well represent sub-par execution quality. In a recent study Plexus analyzed the typical costs incurred by manager style. Exhibit 2 shows the average cost as well as the range of costs from style to style.10
Does Manager Style Affect Costs, and If So What Does Each Style Cost?
Not surprisingly, manager style directly impacts the level of costs incurred to implement investment decisions. As such, the issue is not so much the absolute costs, but rather, whether the manager paid more than expected given the investment style they are hired to implement. Mechanically, a number of elements can make a trade more or less difficult. The primary elements that govern relative trade difficulty are: Order size and liquidity. Large orders (on both an absolute and relative basis) cost more than small orders.8 For this reason it is crucial that manager execution costs be kept consistent with the size of their assets under management (managing $3 billion is quite different from managing $30 billion). Stock price movement. This is the probably the single largest factor in determining a trade’s difficulty (and hence, execution costs). Stocks rise when the shares demanded in the market exceed the number of shares offered to be sold. In such circumstances, a buyer faces adverse trading conditions and must outbid the other buyers to complete the trade. Conversely, a person trying to buy a security when everyone else is trying to sell is in a much more favorable situation and can enjoy the benefit of receiving the highest bid from all sellers. The difference in trade costs between trading in such “adverse” and “favorable” markets can range anywhere between 3% and 8% of the stock’s price.9
EXHIBIT 2 Cost Patterns by Manager Style
Manager Style Large Cap Value Large Cap Growth Index/ Passive Small Cap Value Small Cap Growth
24
Manager Timing Costs
Trade Desk Timing
Market Impact
Commission
Missed Trades 11
Total
1
13
8
15
28
65
82
32
21
10
14
159
31
61
25
9
12
138
5
63
40
20
32
160
136
72
57
18
29
312
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
Are My Managers’ Trading Strategies Consistent with Their Investment Styles?
The trading strategy employed by a manager often reflects a confluence of issues and considerations: investment mandates, assets under management, the trading of many accounts at once, satisfaction of soft-dollar commitments or directed brokerage goals (either the plan sponsor’s or other clients of the manager), and the benchmarks used to assess the trade desk. Often a manager is capable of coordinating these competing interests to the mutual benefit of all. However, at other times these considerations may interfere with efficient trading. Accordingly, plan sponsors should endeavor to ensure that trading and investment styles are not at odds. Are My Managers’ Selection of, and Payment to, Brokers Consistent with Best Execution?
As with total execution costs, plan sponsors should review their managers’ commissions and Market Impact costs to ensure that they are selecting brokers that provide quality services. In this respect, the actual commissions and Market Impact costs should be contrasted against a benchmark that only looks at those two cost components and does not include Manager Delay costs.12 Further, in order to estimate what each trade should cost, it is important to take into account the characteristics of what was being traded and the market environment. Again, as with looking at the manager’s total execution costs, broker benchmarks should adjust for the size of each broker’s trade; the liquidity existing at that time; the capitalization of the stock; the market in which the stock was being traded; the trade side; and the price movement of the stock immediately before the broker tried to execute the manager’s order. In this fashion, the benchmark becomes an apples-to-apples comparison, from which the broker’s trading efficiency can be fairly judged. What Services Did Brokers Provide and Were They Worth It?
In addition to execution, brokers provide a variety of other services, for a variety of fees (typically paid for with commissions). Depending on the difficulty of the trade, different levels of execution services are required. Similarly, different research and related services carry varying price tags. Assessing the costs and relative value of these services should be an important component of a plan sponsor’s oversight efforts. SPRING 2001
On the execution side, a common observation is that brokers typically charge higher commission rates to trade harder orders. The supposition is that higher rates compensate the broker for the additional effort needed to execute the trades. The question of course, is whether the extra commissions truly add value: are they justified by savings in the overall execution costs? The answer to this question can be quantified with the proper benchmark. From our experience, it is hard to overstate the importance of commissions as a tool to facilitate efficient executions. While commissions are typically the smallest cost component, they often represent the key to minimizing overall costs. This is true for a variety of reasons. Managers may look to brokers for a variety of execution related services, for example, market information (who’s buying, who’s selling etc.), finding the other side of their trade, providing liquidity when necessary, secrecy, and guaranteeing delivery and settlement. In support of this supposition, Plexus conducted a study which correlated commission costs with overall execution costs. The conclusions were dramatic. In situations involving large or fast moving orders (the most common type of trades that could be termed “difficult”), not only were total execution costs lower for managers who paid higher commissions, but their captured returns were higher as well!13 The reason for this was simple: in markets where stock prices moved quickly (i.e., a lot of managers wanted to buy/sell the same stock at the same time) the challenge became one of capturing as many of the available shares as possible. Trades that were not completed, by definition, generated significant Opportunity costs. Consequently, a manager’s broker had to be properly motivated to provide that manager with the desired securities. In sum, managers use commission dollars to cultivate broker relationships so they’ll be there when needed. The bottom line is that brokers will offer scarce stocks to those managers whose relationship they value most. It should come as no surprise that many managers also use commission dollars to purchase research from brokers (commonly known as “Soft-dollars”). The types of research purchased from brokers varies tremendously from manager to manager and may include: stock-specific analysis, analytical systems, credit analysis, earnings estimates, economic forecasting, industry analysis, interest rate forecasting, investment strategy, proxy/social responsibility services, risk modeling/optimization, stock selection, and technical research. TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
25
The broker typically provides the manager with the services (produced either in-house or by third-parties) in return for which the manager directs trade through the broker. In theory, the broker charges a commission rate slightly higher than it otherwise would have done, to compensate it for the cost of producing the service provided to the manager. This practice is sometimes referred to as “paying up,” and has been legally sanctioned by the SEC under amendments to Section 28(e) of the Securities Exchange Act of 1934. As noted earlier, plan sponsors, industry groups and regulatory agencies are stepping up their review of manager soft-dollar practices. The oversight is primarily intended to discourage abuses (such as sending trades to brokers incapable of providing adequate execution, or purchasing unnecessary or inappropriate services). In addition, some plan sponsors are beginning to review the execution quality of soft-dollar brokers against difficulty-adjusted benchmarks. Where the actual costs exceed the benchmarks, the excess cost is compared to the market-value of the soft-dollar services received by the manager. The table below sets forth the commission ranges typically paid by managers for different broker services (both execution and soft-dollar). • Electronic Crosses • DOT trades • “Vanilla” type trades • Routine block trades • Soft Dollar trades • Large block and difficult agency trades • Partial principal and difficult trades • High-risk principal and basket trades
1-2 cents per share 2-4 cents per share 3-5 cents per share 4-6 cents per share 4-7 cents per share 5-8 cents per share 7-10 cents per share 15-30+ cents per share
What Types of Problems are We Likely to Find?
Review of manager trading practices may turn up a number of additional issues warranting attention. As with some of the earlier questions, these issues tend to fall into one of two categories: inappropriate actions, or inefficient processes. They include: 80:20 rule. Managers often focus on what they do best. So, for example, a large cap manager typically buys large cap stocks and does it well. However, there will be at least some unusual holdings or situations in their portfolio (maybe 10%-15%). If the manager ignores those 26
securities’ unique trading requirements, inordinate costs are likely. It is surprising how frequently 20% of a portfolio is responsible for 80% of the costs. The good news is that the remedy often entails nothing more than bringing the issue to the attention of the manager. Inconsistent trade and investment strategies. As noted earlier, for a variety of reasons, managers may employ trading strategies inappropriate to the types of stocks they are picking. When plan sponsors suspect such a situation, dialogue with the manager is called for. Some fixes require minimal effort (e.g., consolidating orders sent to the trade desk from multiple portfolios), while other require a firmwide shift in how they do business (e.g., low trade completion rate due to trying to beat the average price of the day while stock prices are moving quickly). Bottomline, if the underlying reason remains unchanged, the excessive costs are likely to continue. Excess costs as a result of increased assets under management. Managers who enjoy superior returns often attract new accounts. Unfortunately, while execution costs rise in lockstep with asset size, stock selection ability does not. At some point, stocks might be attractive in the abstract, but unattractive in actuality due to the increase in execution costs associated with orders. Plan sponsors who see this phenomena need to recognize that it represents a systemic problem that won’t go away without the manager taking affirmative steps (such as expanding the manager’s universe of stock holdings, employing accumulative trade strategies, or limiting the number of new accounts). Inefficient use of brokers. A fact of life for participants in the institutional marketplace is that managers typically trade for multiple accounts, and what they perceive to be in the best interests of all clients. As a consequence their selection of brokers is based on a variety of factors, not all of which may be in the best interests of any one specific account. When reviewing broker execution quality, plan sponsors should take steps to assess whether the brokers are in fact providing quality executions and ascertain the manager’s motivation in using them. Inappropriate use of brokers. Although not particularly common, plan sponsors should also review their managers’ selection of brokers to insure that inappropriate firms are not being used. For example, plan sponsors who prohibit managers from using affiliated broker/dealers, or their commissions to purchase soft dollars, should ensure that the manager is complying with the policy. Conversely, plan sponsors who have directed brokerage goals and targets should also monitor compliance. Further,
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
regardless of whether they have a directed brokerage program, plan sponsors should ensure that they are not being disadvantaged by the manager’s efforts to satisfy the requests of other clients. As discussed more fully below, directed brokerage programs, if not constructed prudently, can generate costs that far exceed the benefits. If such practices are identified, the plan sponsor needs to get assurances from the manager that steps will be taken to address the problem. BENEFITS AND APPLICATIONS
Perhaps half of all the plan sponsors we talk to ultimately ask at some point, So what? Yes, we have a legal obligation to monitor trade costs and maybe we can even identify problem areas, but what good is it? We’re not going to tell our managers how to trade. So what are the bottom-line benefits of transaction cost analysis?14
Fair questions all. The answer is multidimensional. Certain benefits are achieved through monitoring, while other benefits arise in those situations where plan sponsors have a more direct role in trading, such as manager transitions and asset rebalancings, directed brokerage programs, and manager searches. How Can Monitoring Help My Management of the Plan Assets?
Identification of inefficient processes (the “bad apples”). As alluded to above, identifying managers with inefficient trading processes can (and should) play an integral part in helping plan sponsors determine when to terminate a manager (or alternatively place them on a “Watch List”). While trading in and of itself clearly doesn’t drive overall performance, it represents one of the systematic processes that make for superior returns. To the extent plan sponsors can gain insight into the systematic efficiency (or inefficiency) of that process, they are better positioned to assess the likelihood of continued underperformance. Conversely, Plexus has seen instances where an underperforming manager was not terminated because he was able to demonstrate that his processes (including trading) were sound. In this fashion, monitoring transaction costs helps avoid terminations that are delayed, reactive, judgmental, and expensive. SPRING 2001
Early warning system (potential future problems). Review of trading costs and the returns associated with those trades may also provide advance notice of potential problems. Because most managers trade less than their full portfolio each quarter, it may take several years before problems manifest to such a degree that the plan sponsor notices them in the overall returns. Not so with traded returns! The traded portfolio, by definition, shows exactly what the manager bought and sold, and how well those decisions performed—immediately. This provides an Early Warning System of things to come. For example, is a manager straying from their agreed upon investment mandate? Are the manager’s total assets under management beginning to be a problem? Has the manager recently hired a new trader (who may or may not utilize an appropriate trading strategy)? On another level, some practitioners have suggested that the traded portfolio represents the dynamic portion of the manager’s overall portfolio (i.e., the primary source of portfolio returns) during the following three or four quarters. Consequently, review of these returns may provide plan sponsors with a sense of the future performance of their portfolios.15 Adherence to policies. As discussed above, where plan sponsors have trading policies in place (whether the focus is soft-dollar, directed brokerage or some other area) monitoring is essential for tracking adherence.16 It is also helpful in terms of justifying (or challenging) the underlying assumptions of the original policy. Manager searches. For all of the reasons plan sponsors monitor their managers, common sense dictates that the insights gleamed through transaction cost analysis has similar relevance in the initial selection of those managers. While everyone acknowledges that past performance is no guarantee of future returns, plan sponsors are often left with little else in choosing a new manager. Transaction cost analysis can provide the plan sponsor with a degree of comfort that they are hiring a sound and rational process. No one claims that efficient trading is the end-all to be-all of manager performance. However, in a world where performance rankings shift dramatically every quarter, efficient trading may very well be the key to top-quartile performance. After all, knowing which finalist consistently paid less than others to trade the same type of securities is something prudent fiduciaries would like to know—before they made their final selection. Bottom-line, monitoring transaction costs help plan sponsors stay on top of things; both in terms of changes within the management firm, as well as future portfolio TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
27
performance. This in turn enables proactive oversight designed to address potential issues before they manifest into real costs. Can I Develop a Directed Brokerage Program Consistent with Best Execution?
Since the early 1980s directed brokerage has played a unique and increasing role in plan sponsor asset management. For purposes of this article, directed brokerage is defined as, “an arrangement whereby an employee benefit plan sponsor requests its money manager, subject to…best execution, to direct commission business to a particular broker/dealer who has agreed to provide services, pay obligations, or make cash rebates to the plan.” As such, direction represents a clear instance where plan sponsors move beyond prudent delegation and oversight, and instead assume responsibility for some of their managers’ investment decision-making authority.17 In light of simplistic estimates of execution costs, this proactive behavior is not surprising. Plan sponsors are often told that all brokerage firms provide the same quality of execution, and managers should therefore be instructed to use those firms who will charge less or rebate a portion of their commissions back to the plan. More to the point, plan sponsors rightfully consider brokerage commissions as assets of their funds and, to the extent trades can be executed for less without impacting execution quality, don’t want to “leave money on the table.” In this context, directed brokerage programs are looked upon as a means of recouping commissions which would otherwise benefit someone else. Unfortunately, directed brokerage if done imprudently may result in significant performance penalties through trade delays and disruption of a manager’s normal execution strategies. The resulting costs often dwarf the direction benefits. The question therefore becomes, at what point is execution quality compromised more than the savings generated through the directed brokerage program? As noted earlier, efficient executions require that managers pay attention to a variety of ever-changing trade characteristics and market conditions. Custom-tailoring their trade strategy as these conditions dictate is essential to minimizing costs. Plan sponsor requests to direct trades to specific brokers inject a new dynamic with little or no reference to what might otherwise be a optimal trade strategy. The central question becomes what, if any, harm is done to portfolio performance as a result of manager compliance with those direction requests. 28
Although managers strive hard to maximize returns for all clients, direction requests frequently place managers in the position of having to confer execution priority between different accounts. Recognizing how this dynamic works is the first step in developing rational programs. Most large managers have multiple institutional clients with the same investment mandate. Managers also have a fiduciary responsibility to achieve best performance for every account. Consequently, when a manager determines to buy or sell particular security for one client, it often wants to buy or sell for all. In these instances, the manager will aggregate the separate plan sponsor orders into one large “block” order. The block order is then sent to a single brokerage firm to be executed together. Upon execution by the broker, the proceeds and costs are divided among the participating plan sponsors pro rata. In this fashion, no client is disadvantaged with respect to another, and the manager satisfies its fiduciary duty to all. This all changes when one of the accounts instructs the manager to use a different broker. Brokerage firms can vary substantially in terms of financial strength, trading skills, areas of expertise, and so on. Managers may therefore have legitimate concerns that directing trades to new and unfamiliar brokers will limit the manager’s ability to select brokers who: • Are knowledgeable in the manager’s type of stocks; • Know where the sellers are; • Are able to “principal” a trade; • View their relationship as important; • Have proven fidelity; and • Are rewarded for a job well done. In response to these considerations managers have, not surprisingly, engaged in certain practices designed to minimize possible negative consequences to other, nondirected accounts. Specifically, managers often pull directed trades from the rest of the block order to be traded separately from the accounts that did not direct. Since managers have an obligation to seek the best possible price for the greatest number of clients, they tend to place (“sequence”) the remaining block of aggregated orders in front of directed-trades (which would have been part of the block order but for the plan sponsor’s direction). In practice, this means that the manager will wait until the block order is completed before even beginning to try and execute the directed order. Should the price of the stocks being bought/sold move rapidly (as is often the case with
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
EXHIBIT 3 Directed Commissions and Commission Recapture Where It Works
Easy trades that require simpler execution facilities:
• • • •
Where It Doesn’t Work
Difficult trades, often generated by managers who specialize in fast moving ideas:
reasonable order size (less than 25%- • large order size (over several day’s 50% of average daily volume); volume); larger capitalization companies; • small capitalization companies; and exchange-listed companies; and • adverse price momentum. during quiet markets (not a lot of adverse price momentum).
growth or momentum style managers), the delayed directed trades can miss out on a significant amount of the stock’s price appreciation. Further, as more plan sponsors direct brokerage, splitting up block orders results in longer queues, which in turn increasingly slow execution. A recent, and arguably less onerous, alternative to “sequencing” directed trades is a procedure called “step outs.” Mechanically, rather than pulling a client’s order out of a block order to be sent to that client’s directed broker, the manager would negotiate with the broker executing the block order to “give up” to the directed broker a portion of commissions attributable to the directing plan sponsor. Although the manager’s broker, in effect, executes the directing plan sponsor’s trades for free, most brokers are willing to accommodate such requests if the percentage to be stepped out is not too high. In this manner, the plan sponsor’s direction request is satisfied without having to delay the actual implementation of its trades. Plexus has conducted numerous studies (most recently in the Summer of 1998), all of which point to the same general conclusions: on average, directed trades increased trading costs by 16 cents and reduced returns by another 4.5 cents per share. In comparison to nondirected accounts, while directing accounts saved 3 cents in recaptured commissions they lost 10.3 cents in performance.18 However, in those instances where stock prices remained fairly neutral and time delays therefore not crucial (typically seen with large cap value-oriented managers) the difference disappears. SPRING 2001
Bottom-line, directed brokerage can be conducted consistent with fiduciary standards of prudence, but plan sponsors would be well advised to follow two rules of thumb when doing so. Easy trades typically represent between 10%-30% of a manager’s trading (depending on the manager’s investment mandate and amount of assets under management). As with easy trades, difficult trades typically comprise about 10%-30% of a manager’s trading, leaving 40% of medium difficulty, which may or may not be amenable to directed trading (Exhibit 3). After careful consideration of each manager’s investment and trading styles, plan sponsors must ultimately decide whether to direct brokerage; and if so, how much and through whom. 1. Plan sponsors should limit direction to a modest portion of each managers’ commissions, typically 10%-30% (depending on the mix of managers); 2. Plan sponsors should choose directed brokers carefully in consultation with their managers (managers are best positioned to assess broker skills and payment mechanisms); 3. Plan sponsors should devise direction targets on a manager-by-manager basis recognizing that diversified and value managers are generally best positioned to deal with directed brokers because their decisions are less time sensitive; 4. Plan sponsors should encourage their managers to explore the use of step outs rather than sequencing when satisfying direction goals; TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
29
5. Since managers typically use the commissions from easy trades to purchase soft dollar services, plan sponsors should consider whether these services are essential to manager performance; and 6. Plan sponsors should install a monitoring program to verify continuing execution quality and performance. Manager Transitions (and Asset Rebalancing) —What Do I Need To Know?
From time to time, plan sponsors make major asset shifts resulting in assets being taken from, or directed to, individual managers. In these instances, the plan sponsors play a primary role in determining the costs associated with those transitions. Whether by design or default, plan sponsors establish parameters and policies that, to a large degree, predetermine their level of transaction costs. For example, a plan sponsor, in determining how to liquidate a portfolio, must decide whether they will consider bids on both a principal and/or agency basis. The fact that a plan sponsor only considers agency bids is an implicit decision not to consider principal bids. Depending on the characteristics of the portfolio to be liquidated and market conditions at the time of the transition, the plan sponsor may thereby miss an opportunity for significant cost savings. While it has occasionally been suggested that transition costs, in comparison to managers’ overall performance, represent only “nickels and dimes,” these excess costs can easily range upwards of several hundred basis points and hundreds of thousands of dollars.19 Other than the establishment of asset allocations and selection of managers, there are few instances where plan sponsors have such direct impact on the assets of their funds. What Should I Consider in Developing a Transition Strategy?
As a general rule, transitions begin with the plan sponsor’s determination that a certain portion of the fund’s assets have to be repositioned; either to cash, another manager, or a different asset class or style. A transition strategy can be viewed as a series of questions, the answers to which tend to point the plan sponsor towards a particular direction and type of transition agent. The questions include: 30
Transition time horizon and external time constraints. Small liquid transitions can be accommodated virtually instantly, while larger transitions may require one day to several weeks or more to complete. In addition, plan sponsors may have specific time constraints within which the transition must be completed. For example, the sponsor may need to raise cash immediately to pay benefits or operational expenses. Minimizing total costs versus commissions or marketimpact. While the costs that a plan sponsor pays consist of three components: commissions, market-impact and delay costs, a plan sponsor may have priorities in addition to the general goal of minimizing aggregate costs. For example, the plan sponsor may consider the potential market-impact and delay costs to be inconsequential (in comparison to its budgetary needs) and therefore focus on minimizing commissions through a commission recapture broker. Assumption of market risk. Ultimately, a plan sponsor’s cost priorities and sensitivities will influence his selection of a transition agent and compensation arrangement. A Plan Sponsor who is comfortable assuming the risk of large market-impact and delay costs may choose an agency bid. However, given that agency bids only guarantee commissions, Plan Sponsors considering agency bids ought to have a sense of the potential market-impact and delay costs (often incurred through no fault of the broker, but rather simply as a function of market conditions prevailing at that time). If the plan sponsor is reluctant to take on such unknown risks, a principal bid may be more appropriate. As discussed later, Principal brokers actually purchase the entire transition portfolio from the plan sponsor. Accordingly, any market-impact and delay costs subsequently incurred, accrue to the broker who now owns the portfolio rather than the plan sponsor. Of course, principal brokers are not in the business of taking other people’s risk without getting paid for it, which is why principal bids are typically higher than the commissions quoted in the agency bids. A third alternative would be to retain an agency broker pursuant to an incentive fee arrangement. Rather than simply agree to a commission level, a plan sponsor may structure the fee arrangement so as to incent the broker to minimize market-impact and delay costs. In these instances, care should be taken that the benchmark does in fact align the interests. For example, a commonly used benchmark in these arrangements is the daily volume weighted average price (the VWAP). Unfortunately, such a measure only looks at market-impact, but
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
not delay costs. Using such a benchmark could incent a broker to minimize market-impact, even if the delay costs dwarfed any bottom-line savings to the fund. End result: the fund’s market-impact costs would indeed be low (and in all likelihood, the broker would therefore qualify for its additional fee), but the overall transition costs might be much higher due to the delay costs incurred; particularly in transitions of more than one day. Market exposure (derivatives and other strategies). Given that a portfolio of stocks will generally outperform a portfolio of cash, plan sponsors should minimize the amount of time during which portfolio assets are liquidated but not reinvested. Throughout that interim period, the assets are, in effect, providing returns equal to the returns on cash rather than stocks. With assets invested in cash rather than equities the transition portfolio is forgoing very real investment returns. This phenomenon is known as “cash drag.” The insidious cost of cash drag can be illustrated with a simple example. In 1997, bottom-quartile equity managers generated returns of up to 22.9% while Treasury Bills yielded 5.29%. If a plan sponsor terminated a bottomquartile manager and the transition took only two weeks, the cost of being invested in cash verses bottom-quartile performance would have been over 67 basis points! One method for maintaining market exposure is to use derivatives such as futures. Although derivatives carry a small fee, the amount is virtually nil when compared to the potential costs of not maintaining market exposure. If the plan sponsor is unable or unwilling to use such instruments, alternative strategies are available. The plan sponsor can instruct an agency broker to utilize “dollar neutral” strategies. In these instances, the broker orchestrates the transition so that it simultaneously liquidates a security and buys a replacement. In this manner, there is virtually no time in which the portfolio is invested in cash. The challenge to the broker is to minimize that time lag; and the challenge to the plan sponsor is to select a broker with sufficient back-office technologies to do the best job. Who Should I Select As My Transition Agent?
After thoroughly analyzing the transition portfolio and determining an appropriate strategy, the plan sponsor must then decide who should be retained to implement the strategy. The plan sponsor has several options, each of which brings strengths and weaknesses:
SPRING 2001
1. Active money managers (the terminated manager, new manager, or a manager unrelated to the transition); 2. Large passive index funds (often affiliated with the plan sponsor’s custodial bank); 3. Agency brokers (either full service brokers, commission recapture/discount brokers, or brokerage arms affiliated with the plan sponsor’s investment consultant or custodial bank); and 4. Principal broker/dealers. The systematic strengths and weaknesses of each type are described below. Terminated manager. One option is to simply instruct the old manager to unwind their positions into cash. Certainly the old manager is most familiar with the stocks in their portfolio and positioned to liquidate them efficiently. However, several serious issues are raised by this strategy. The plan sponsor needs to be concerned about the diligence the terminated manager will devote to executing those trades. Not that the manager will purposely disadvantage the portfolio (the manager is still a fiduciary after all). Rather, will the manager go the extra mile in pursuing optimal execution, or will trades be used to pay off soft-dollar obligations regardless of the soft-dollar broker’s ability to provide best execution? New manager. Another common alternative is to simply authorize the new manager to sell the old portfolio and reinvest the proceeds as it sees fit. In terms of liquidating the old portfolio, the new manager is often given an opportunity to pick any securities from the old portfolio that they would like to have (informally referred to as “cherry picking” or “in-kind transfers”). In-kind transfers offer a cost-free method of transferring those names. In point of fact, unless there are specific reasons limiting the disbursement of the liquidated portfolio, all of plan sponsor’s existing managers should be given an opportunity to cherry-pick. This represents the single most cost effective transition strategy. After “cherry picking” from the old portfolio, the new manager will then sell off the remaining stocks. As with the terminated manager, plan sponsors should monitor the new manager to ensure those trades are not used to satisfy soft-dollar commitments without regard to the execution quality. This is not usually a serious concern, since in most instances managers will be more interested in ensuring that the new relationship gets off on the best foot possible. A little awareness can avoid disappointment, however. TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
31
More importantly, plan sponsors should verify that the new manager’s trade desk has the expertise to effectively trade those stocks. A manager’s trading desk will always be most familiar with the stocks they trade on a daily basis. They know which brokers have access to liquidity in those names, how best to trade them, and so on. However, if unfamiliar with the holdings in the old portfolio, the new manager though willing, may not be capable of providing the most efficient liquidation. Clearly, this issue grows in significance as familiarity decreases. In the event the new manager is chosen to liquidate the old portfolio, they typically request that the reinvestment period not be considered part of its investment performance. While this request is not without merit (since it entails an abnormal type of trading with little relevance towards normal day-to-day activity), the plan sponsor should insist that the reinvestment period be monitored as a stand-alone event; with the manager’s regular investment performance beginning upon completion of the reinvestment. Large index funds. Many plan sponsors utilize large passive managers (often affiliated with the sponsor’s custodial bank) to help in liquidating transition portfolios. In theory, these index funds are able to cross many of the stocks in the transition portfolio (thereby incurring no marketimpact on those trades). While this can often be a very cost efficient practice, there are limitations to this strategy. First and foremost, plan sponsors should ascertain how frequently the index fund crosses stocks. Infrequent crosses can have the effect of maintaining the old portfolio for an extended period of time which, may result in significant costs that an active transition agent would have avoided. Further, index managers are rarely, if ever, able to cross the entire transition portfolio. This leaves the remaining portfolio to be traded by the index manager’s trade desk. While the index manager may argue this is no different than any other transition agent, the index manager’s trade desk often has a passive orientation. As a consequence, they may have neither the technical tools nor inclination to aggressively seek out liquidity, which could be necessary to trade the remaining securities in an efficient manner. A third, more insidious cost arises from the fact that an index fund’s primary emphasis is to maximize crossing opportunities, thereby minimizing market-impact costs. In any large portfolio of stocks, certain names will be in demand while others are not. Under normal circumstances, a broker would expect to pay a little extra to 32
entice someone to take an unwanted stock (known within the industry as “paying for liquidity”). Conversely, a broker would expect to pay a little less when trading a stock that everyone else wanted (i.e., as “providers of liquidity” they would receive a premium). As a general rule, the very stocks for which an aggressive agency broker could have received price concessions get crossed at no market gain. For this reason, plan sponsors should consider the profile of the transition portfolio in determining the optimal strategy. Stocks with favorable price swings (i.e. rising sells and falling buys) may represent opportunities for marketgains if actively traded (gains that would help offset the cost of trading difficult stocks). Agency brokers. Agency brokers do not use their own capital to purchase the customer’s orders themselves, but rather act as a middleman between the buyer and seller. As such, agency brokers commit only to a set commission rate that is paid on each trade. While the broker implicitly commits to making its “best efforts” to keep delay and market-impact costs to a minimum, any costs so incurred are paid by the plan sponsor out of the assets of the transition portfolio. So for example, an agency broker, with an order to buy 50,000 shares of a stock selling for $50 might charge 6 cents per share commission. If the broker ultimately executes the order at $49.50, the plan sponsor, pays 6 cents per share ($3,000) for commissions and 50 cents per share ($25,000) to market-impact and delay costs. All three components are imbedded in the price ultimately paid for the stock, and come out of the assets of the Fund. Principal broker/dealers. Unlike agency brokers, principal brokers (also known as dealers) use their own money to actually buy (or sell) the stocks from (or to) a plan sponsor. Using the earlier example of an agency broker trying to sell 50,000 shares of a $50 stock, if a principal broker submitted a bid of 20 cents per share, the 20 cents would constitute the entire costs. Any marketimpact or delay cost subsequently incurred would be the broker’s costs. In this example then, the plan sponsor’s total costs would be $10,000 (50,000 shares 20 cents per share). Of course, since the transition agent cannot predict exactly the market impact and delay costs, their bid typically takes into account the additional risks they are taking on. This is why principal brokers often refer to principal bids as “risk” bids. Since the principal broker now owns the transition portfolio, they inherit the risk of marketimpact and delay costs from the plan sponsor. Principal brokers are particularly eager to bid on plan sponsor transitions, since unlike manager-initiated trades, plan sponsor transi-
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
tions are not driven by new/secret information, which may dramatically affect the price of the stocks. Mechanically, a plan sponsor considering principal bids would send RFPs (often in the afternoon of the day they wanted to complete the transition) to two or three brokers. The RFPs would typically not disclose the actual names of the stocks to be bought and sold (for fear that the bidders could use that information in a manner that adversely affected the value of the portfolio), but rather descriptive characteristics of the portfolio as a whole (commonly known as “blind bids”). Competing brokers typically submit their bids within an hour of receiving the RFP. The winning broker is then sent the list of the securities to be liquidated and purchased. In return, the plan sponsor receives either a check in an amount equal to the portfolio value less the bid price (if the transition was just a liquidation), or the securities identified on the new manager’s buy list. At that point the transition is completed. What Capabilities Should I Look for in a Transition Agent?
Just as no two transition portfolios are exactly alike, no two agency brokers have the same strengths and weaknesses. The challenge for the plan sponsor then, is to match the capabilities of the broker to the needs of the transition portfolio. For this reason, agency brokers should be able to demonstrate specific capabilities in several areas: Execution expertise with respect to the characteristics of the transition portfolio. If, for example, the portfolio consists mainly of small cap stocks, the plan sponsor should ensure the broker has access to, and an intimate knowledge of, those markets. Access to multiple sources of liquidity. As discussed earlier, liquidity is often the key to controlling trading costs. Transitions with significant percentages of illiquid orders require agency brokers with access to multiple sources of liquidity. These sources include: membership on all key domestic and international stock exchanges, a strong presence in portfolio trading and block desk capabilities, direct access to all electronic crossing networks (i.e., Instinet, POSIT, Bridge etc.), direct access to automatic exchange routing systems (i.e., Super DOT), index funds, other institutional desks and market makers, and substantial customer order flow.20 Strong pretransition analytics. In complicated transitions (where for example, the broker is being asked to maintain sector neutrality) pretransition analysis is essenSPRING 2001
tial in establishing the sequence and techniques by which the separate portions of the transition portfolio are traded. Accordingly brokers should have the tools necessary to perform similar analysis. Strong post-transition analytics. Requiring the broker to provide a final transition assessment hightens their awareness (and therefore the likelihood) that execution efficiency must be attained. Strong trading support capabilities. Brokers should be able to demonstrate their ability to monitor and track the transition during actual implementation. This is particularly important in complicated transitions where a broker utilizes multiple trading mechanisms and sources of liquidity. Oftentimes, the concurrent use of crossing networks and the floors of exchanges inadvertently generates temporary imbalances in the transition portfolio’s exposure to specific market sectors or industries. The broker must have the internal capability to identify and correct imbalances as quickly and cheaply as possible. Quality of personnel and reputation for integrity, trust, and effectiveness. While more difficult to quantify, plan sponsors should ascertain the transition agent’s reputation within the industry. References from past clients who conducted transitions similar in nature to the plan sponsor, can provide information on: 1. The characteristics of their transition portfolio and transition strategy employed; 2. Any unanticipated issues that arose and how they were addressed by the transition agent; 3. The effectiveness, efficiency, and creativity of the transition agent; 4. The plan sponsor’s ultimate satisfaction with the transition, and 5. Any measures of quality of execution. Given that principal bids represent the full cost of a transition, assessing a principal broker’s capabilities is less important than with agency brokers (whose internal strengths and weaknesses will largely determine the level of cost incurred by the plan sponsor). Successful principal brokers typically have strong technology capabilities (which enable them to manage and offset the risks they assume by purchasing the transition portfolio), a tolerance for risk-taking, knowledge of the market, and access to capital. Today even large portfolios often receive bids comparable in price to agency bids; although the universe of principal brokers willing to submit serious bids on portfolios over $1 billion is limited to a few of the largest firms. TRANSACTION COSTS
It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
33
How Should I Monitor the Transition?
Although most transition agents offer (indeed, should be required to provide) post-transition analysis, a plan sponsor may choose to obtain an unbiased independent analysis. The post-trade analysis should assess the overall efficiency of the transition and serve as a reference point for future transitions. At a minimum, the summary should quantify the beginning and ending portfolio values, the number of shares traded, and the breakdown of costs versus agreedupon benchmarks. Many summaries also quantify each day’s beginning and ending portfolio values, and stock-specific analysis. As discussed earlier, it is also important that plan sponsors utilize an appropriate benchmark in assessing the efficiency of their transition agents, and communicate that choice to the agents. There are a variety of benchmarks available, not all of which measure the same things. For example, a daily VWAP-type benchmark, may often be used to assess the market-impact and commission costs, but would be inappropriate to assess the total transition costs since it does not capture the delay costs associated with the transition. Furthermore, daily average measures, such as the VWAP, are in essence static yardsticks that judge each trade against the same standard (without regard to difficulty). As a result, counter-intuitive assessments such as a manager who paid 10 cents per share to sell 1,000 shares of a stock being judged as superior to a manager that traded 1,000,000 shares of the same stock for a penny more. To address this anomaly, the most relevant comparisons are benchmarks that account for the specific characteristics of what was being traded and the market conditions in which they were being traded (i.e., a “difficulty-adjusted” benchmark).
How Should I Interact with the Manager if a Problem Is Spotted?
Concurrent with instituting a monitoring program, plan sponsors are well served to initiate a dialogue with their managers to obtain a better understanding of their trading practices. This will not only assist in subsequent monitoring efforts, but also help ensure that plan sponsor-initiated programs do not unintentionally impact performance as well. Relevant questions to ask managers include:
CONCLUSION
In summary, transaction costs are an important matter for plan sponsors, managers and brokers alike. To satisfy their fiduciary obligation to ensure best execution, plan sponsors should be prepared to take a hard look at their manager’s trading practices. While this is not necessarily an easy task, instituting a sound monitoring program is a good place to start.
34
Manager Trading Questionnaire
• What are your total execution costs? What is the breakdown among the various cost components? What does it buy for the portfolio? • What does the manager consider to be reasonable execution costs given their investment mandate? Under what conditions would the manager be willing to pay higher costs? How often do those conditions arise? • How does the manager coordinate its trading style with its investment style? What tradeoffs (between the different cost components) does the manager consider when defining their trade strategy? • How does the manager communicate these sensitivities to the brokers it utilizes? Does that communication change the brokers’ behavior? • How does the manager decide what portion of the commissions generated by its trades goes to brokers from which it expects good execution on difficult trades, or to brokers from which it buys research and soft dollars? How does the manager know when the ratio is out of balance? • What does the manager do to assure that a broker is providing “Best Execution?” What does the manager do when it senses there is a problem? • How does the manager validate/measure it own trade desk’s abilities? How are their traders compensated? What is the manager doing to improve its implementation of investment ideas?
WHAT EVERY PLAN SPONSOR NEEDS TO KNOW ABOUT TRANSACTION COSTS Copyright © 2004 Institutional Investor, Inc. All Rights Reserved
SPRING 2001
Plan sponsors need to know what they pay to buy and sell securities in their portfolios and whether those costs were reasonable. Experience has shown that attention to the transaction process typically produces risk-free, recurring contributions to client returns. In contrast, inattention to these issues results in slow, continuing leakage of performance. As Carole Ryavec, a managing director of Plexus Group has said, “Ten basis points, lost over and over again, is a lot of money.” ENDNOTES This article first appeared in the Journal of Investment Consulting published by the Investment Management Consultants Association (IMCA). 1 “Picking Good Stocks: Necessary But Sufficient?— Commentary #43,” Plexus Group (January 1995). For further information of Plexus and services offered to the institutional investment community, visit our web site at: www.plexusgroup.com. 2 Reflecting roughly 25% of exchange volume. 3 Based on the 10-year returns through June 30, 1998. 4 Assuming a turnover rate of 100%. 5 “Pay Me Now Or Pay Me Later: The Interplay of Impact, Timing and Opportunity—Commentary #32,” Plexus Group (September 1991). 6 Although the coverage of ERISA does not technically extend to public plan sponsors, courts in all likelihood would look to it in deciding the merits of similar fact scenarios. 7 Indeed, in December 1997 the Association for Investment Management and Research (“AIMR”) published new soft-dollar standards in which it required managers to make available detailed information regarding their soft-dollar practices. This information included:
8
Wagner, Looking Below the Iceberg Tip, Plan Sponsor, April 1998. 9 Id. 10 “How Manager Style Influences Cost—Commentary #55,” Plexus Group (May 1998). 11 Portfolio effect; adjusted for rate of completion. 12 “The Official Icebergs of Transaction Costs—Commentary #54,” Plexus Group (January 1998). 13 “In Defense of the Commission—Commentary #40,” Plexus Group (April 1994). 14 These plan sponsors often see no contradiction in directing trades to certain brokers to procure benefits for their funds. 15 “Identifying A Quality Investing Process—Commentary #48,” Plexus Group (May 1996). 16 “Soft Dollar Arrangements,” SEC Rel. No. 34-23170 (April 23, 1986). 17 Indeed, some practitioners estimate that between 40%-50% of all plan sponsors currently direct at least some brokerage. See, e.g., K. Kahn and B. Leonard, separate lectures at Information Management Network Public Fund Boards Summit (March 16, 1997). 18 “Revisiting Directed Brokerage: Still No Free Lunch —Commentary #57,” Plexus Group (December 1998). 19 See, e.g., Plexus Group, Plan Sponsor Case Study: Transition Blues. 20 Particularly with smaller cap securities, the fact that a broker is a “market maker” in a lot of names can be a significant advantage in maximizing the opportunities of crossing what otherwise might be expensive trades.
• Descriptions of each product and service received; • Identification of the producers of each product and service received; • The aggregate commissions paid from the client’s account to each broker; • The aggregate capital value of all principal trades sent to each broker; • The aggregate percentage of the manager’s brokerage derived from client-directed arrangements; • The amount of brokerage directed by the client to each broker (as both an absolute number and as a percentage of the aggregate brokerage directed by all of the manager’s clients); and • The aggregate amount of brokerage paid from all of the manager’s accounts.
SPRING 2001
TRANSACTION COSTS It is illegal to reproduce this article in any format. Email
[email protected] for Reprints or Permissions.
35
